Michellamine B (MB; NSC 649324) is a naturally occurring naphthylisoquinoline alkaloid with activity against human immunodeficiency virus (HIV) types and 2 in vitro. In conjunction with its preclinical evaluation, the plasma pharmacokinetics of MB was characterized in mice and beagle dogs treated by intravenous (iv) infusions of 1 and 15 min duration, respectively. At doses ranging from 1 to 9 mg/kg, the drug exhibited apparent first-order kinetics in both species, affording triexponential plasma concentration-time profiles. Peak plasma levels of MB within the range that completely inhibits the cytopathic effects of HIV upon cultured human lymphoblastoid cells (50-100 micro g/mL) were achieved without evidence of toxicity. MB had a biological half-life of 2.8 +/- 0.8 hr in mice, with a 2.1 +/- 0.3 hr mean residence time and a total plasma clearance of 2.4 +/- 0.5 mL/min/kg (mean +/- SD, n=3 studies); however, the terminal phase contribution to the area under the plasma profile from time zero to infinity was only 44.6 +/- 12.9%. In contrast, the terminal phase was the primary determinant of drug disposition in dogs, accounting for 74.1 +/- 2.8% (n=3) of the area under the curve. The systemic duration of MB was significantly longer in the dog, as indicated by mean values of the apparent half-life (11.6 +/- 1.2 hr), mean residence time (12.3 +/- 1.8 hr) and clearance (0.50 +/- 0.08 mL/min/kg). However, there was no statistical difference between the apparent volume of distribution in the mouse (0.60 +/- 0.08 L/kg) and dog (0.50 +/- 0.07 L/kg), indicative of similar binding characteristics of the drug with plasma proteins in relation to tissues. The relative ease of drug elimination may be attributed to interspecies differences in intrinsic clearance. In a dog given a total dose of 97 mg/kg as a 72 hr constant rate iv infusion, the MB plasma concentration exceeded 20 micro g/mL at 4 hr and gradually increased to 41 micro g/mL at the end of infusion. There were no clinical or pathological indications of toxicity. The total plasma clearance (0.48 mL/min/kg) was within the range observed in dogs treated by 15 min infusion, extending the postinfusion sampling period from 24 hr to 4 days facilitated better definition of the terminal exponential phase, yielding a value of 25.6 hr for the biological half- life of MB. The amount of drug excreted by dogs unchanged in the urine ranged from 3.7 to 11.1% of the administered dose. The major pathways by which the drug is eliminated from the body remain to be identified. On the basis of these findings, continued preclinical development of this chemotherapeutic agent for the treatment of HIV infection is warranted.